R/simluate.R

In thibautjombart/fancysamplers: fancy ways to sample from complicated densities

## #' Simulate multimodal univariate distribution
## #'
## #' Simulate multimodal univariate distribution using a mixture of 'k' normal distributions.
## #'
## #' @export
## #'
## #' @rdname multimod
## #'
## #' @aliases rmultimod dmultimod
## #'
## #' @author Thibaut Jombart \email{t.jombart@@imperial.ac.uk}
## #'
## #' @param n a number of values to simulate
## #' @param k a number of modes for the distribution
## #' @param mean a vector of means (optional)
## #' @param sd a vector of standard deviations (optional)
## #' @param mean.range the range of the randomly generated means
## #' @param sd.range the range of the randomly generated standard deviations
## #'
## #' @details This is essentially a wrapper for rnorm with different default values
## #'
## #' @examples
## #' set.seed(1)
## #' x <- rmultimod(100)
## #' hist(x, col="grey", nclass=50, border="white")
## #'
## #' @importFrom stats rnorm runif
## #'
## rmultimod <- function(n, k=2, mean=NULL, sd=NULL,
##                       mean.range=c(0,100), sd.range=c(1,1)){
##     ## GET VECTOR OF MEANS ##
##     if(is.null(mean)){
##         mean <- runif(k, min=min(mean.range), max=(max(mean.range)))
##     }

##     ## GET VECTOR OF SDS ##
##     if(is.null(sd)){
##         sd <- runif(k, min=min(sd.range), max=(max(sd.range)))
##     }

##     ## GENERATE VARIABLES ##
##     out <- rnorm(n, mean=mean, sd=sd)

##     ## RETURN ##
##     return(out)
## } # end rmultimod




## #' @rdname multimod
## #' @export
## #'
## dmultimod <- function(x, k=2, mean=NULL, sd=NULL, log=TRUE,
##                       mean.range=c(0,100), sd.range=c(1,1)){
##     ## GET VECTOR OF MEANS ##
##     if(is.null(mean)){
##         mean <- runif(k, min=min(mean.range), max=(max(mean.range)))
##     }

##     ## GET VECTOR OF SDS ##
##     if(is.null(sd)){
##         sd <- runif(k, min=min(sd.range), max=(max(sd.range)))
##     }

##     ## COMPUTE DENSITIES ##
##     out <- double(length(x))
##     for(i in 1:k){
##         out <- out + dnorm(x, mean=mean[i], sd=sd[i] , log=log)
##     }

##     ## RETURN ##
##     return(out)
## } # end dmultimod